From 424d55f3a3164dabcfb8cdf9fe6cd8b7d339b9d5 Mon Sep 17 00:00:00 2001
From: Greg Sabino Mullane <greg@turnstep.com>
Date: Sat, 31 May 2025 00:38:26 -0400
Subject: [PATCH] Allow specific information to be output directly by Postgres.

---
 doc/src/sgml/config.sgml                      |  78 +++++++
 src/backend/tcop/backend_startup.c            | 191 ++++++++++++++++++
 src/backend/utils/misc/guc_parameters.dat     |  18 ++
 src/backend/utils/misc/postgresql.conf.sample |   8 +
 src/include/postmaster/postmaster.h           |   4 +
 src/test/modules/test_misc/meson.build        |  1 +
 src/test/modules/test_misc/t/009_expose.pl    | 91 ++++++++++++++++++++++
 7 files changed, 391 insertions(+)
 create mode 100644 src/test/modules/test_misc/t/009_expose.pl

diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml
index e9b420f3ddb..a3d4b388856 100644
--- a/doc/src/sgml/config.sgml
+++ b/doc/src/sgml/config.sgml
@@ -1077,6 +1077,84 @@ include_dir 'conf.d'
      </variablelist>
      </sect2>
 
+     <sect2 id="runtime-config-expose-settings">
+     <title>Expose Settings</title>
+
+     <variablelist>
+
+     <varlistentry id="guc-expose-recovery" xreflabel="expose_recovery">
+      <term><varname>expose_recovery</varname> (<type>boolean</type>)
+      <indexterm>
+       <primary><varname>expose_recovery</varname> configuration parameter</primary>
+      </indexterm>
+      </term>
+      <listitem>
+       <para>
+        Enables reporting if the server is in recovery mode without requiring
+        an authenticated login. Clients can send the string <literal>GET /replica</literal>
+        and will receive a 1 or 0. This is equivalent to logging in and running
+        <literal>SELECT pg_is_in_recovery()</literal>. A client can also send the
+        string <literal>HEAD /replica</literal> which will solely return an HTTP literal:
+        <literal>200</literal> if the server is in recovery, <literal>503</literal> if not.
+        (This allows a drop-in replacement to the same Patroni functionality)
+        Finally, a client can issue <literal>GET /info</literal> and receive the string
+        <literal>RECOVERY: </literal> followed by a 1 or 0.
+       </para>
+      </listitem>
+     </varlistentry>
+
+     <varlistentry id="guc-expose-sysid" xreflabel="expose_sysid">
+      <term><varname>expose_sysid</varname> (<type>boolean</type>)
+      <indexterm>
+       <primary><varname>expose_sysid</varname> configuration parameter</primary>
+      </indexterm>
+      </term>
+      <listitem>
+       <para>
+        Enables reporting the system identifier of the cluster without requiring
+        an authenticated login. Clients can send the string <literal>GET /sysid</literal>
+        and will receive the numeric system identifier. This is a unique number generated
+        by each cluster when initdb is run.
+       </para>
+       <para>
+        A client can issue <literal>GET /info</literal> and receive the string
+        <literal>SYSID: </literal> followed by the numeric system identifier.
+       </para>
+       <para>
+        This feature is useful for determining if the server is the same server as previously
+        encountered. Note than primary and replica servers will share the same system
+        identifier.
+       </para>
+      </listitem>
+     </varlistentry>
+
+     <varlistentry id="guc-expose-version" xreflabel="expose_version">
+      <term><varname>expose_version</varname> (<type>boolean</type>)
+      <indexterm>
+       <primary><varname>expose_version</varname> configuration parameter</primary>
+      </indexterm>
+      </term>
+      <listitem>
+       <para>
+        Enables reporting the numeric version of the Postgres cluster without requiring
+        an authenticated login. Clients can send the string <literal>GET /version</literal>
+        and will receive an integer representing the version.
+       </para>
+       <para>
+        A client can issue <literal>GET /info</literal> and receive the string
+        <literal>VERSION: </literal> followed by the numeric version.
+       </para>
+       <para>
+        This is particularly useful for non-Postgres systems (esp. security scanners) that
+        need a way to easily determine the version of Postgres in use without requiring
+        a Postgres client - or without needing any knowledge of the Postgres protocol at all.
+       </para>
+      </listitem>
+     </varlistentry>
+
+     </variablelist>
+     </sect2>
+
      <sect2 id="runtime-config-connection-authentication">
      <title>Authentication</title>
 
diff --git a/src/backend/tcop/backend_startup.c b/src/backend/tcop/backend_startup.c
index 14d5fc0b196..c69a0727693 100644
--- a/src/backend/tcop/backend_startup.c
+++ b/src/backend/tcop/backend_startup.c
@@ -46,6 +46,10 @@
 bool		Trace_connection_negotiation = false;
 uint32		log_connections = 0;
 char	   *log_connections_string = NULL;
+bool		expose_recovery = false;
+bool		expose_sysid = false;
+bool		expose_version = false;
+
 
 /* Other globals */
 
@@ -65,6 +69,7 @@ static void SendNegotiateProtocolVersion(List *unrecognized_protocol_options);
 static void process_startup_packet_die(SIGNAL_ARGS);
 static void StartupPacketTimeoutHandler(void);
 static bool validate_log_connections_options(List *elemlist, uint32 *flags);
+static bool ExposeInformation(pgsocket fd);
 
 /*
  * Entry point for a new backend process.
@@ -148,6 +153,15 @@ BackendInitialize(ClientSocket *client_sock, CAC_state cac)
 	StringInfoData ps_data;
 	MemoryContext oldcontext;
 
+	/*
+	 * Scan for a simple GET / HEAD request. If this is detected and
+	 * handled, we are done and can immediately exit
+	 */
+	if ((expose_recovery || expose_sysid || expose_version)
+		&& ExposeInformation(client_sock->sock))
+		proc_exit(0);
+	/* Should we do exit(0) here, despite the warnings in ipc.c? */
+
 	/* Tell fd.c about the long-lived FD associated with the client_sock */
 	ReserveExternalFD();
 
@@ -1126,3 +1140,180 @@ assign_log_connections(const char *newval, void *extra)
 {
 	log_connections = *((int *) extra);
 }
+
+
+static
+bool
+ExposeInformation(pgsocket fd)
+{
+
+/*
+ * ExposeInformation
+ *
+ *
+ * Handle early socket probe before full backend startup.
+ * Responds to small set of predefined endpoints (e.g. GET /info)
+ *
+ * Requires at least one "expose_" GUC to be true.
+ *
+ * Returns true if any endpoint is recognized.
+ */
+
+#define EXPOSE_MIN_QUERY 9		/* Shortest possible line: "Get /info" */
+#define EXPOSE_MAX_QUERY 16		/* Longest possible GET line */
+
+/* What information is being returned */
+	typedef enum
+	{
+		EXPOSE_NOTHING,
+		EXPOSE_HEAD_REPLICA,
+		EXPOSE_GET_ALL,
+		EXPOSE_GET_REPLICA,
+		EXPOSE_GET_SYSID,
+		EXPOSE_GET_VERSION,
+	}			ReturnType;
+
+	typedef struct
+	{
+		const char *endpoint;
+		const bool *require;
+		ReturnType	type;
+	}			endpoint_action;
+
+	static endpoint_action endpoint_actions[] =
+	{
+		{
+			"HEAD /replica", &expose_recovery, EXPOSE_HEAD_REPLICA
+		},
+		{
+			"GET /replica", &expose_recovery, EXPOSE_GET_REPLICA
+		},
+		{
+			"GET /sysid", &expose_sysid, EXPOSE_GET_SYSID
+		},
+		{
+			"GET /version", &expose_version, EXPOSE_GET_VERSION
+		},
+		{
+			"GET /info", NULL, EXPOSE_GET_ALL
+		}
+	};
+
+	ssize_t		n;
+	char		buf[EXPOSE_MAX_QUERY + 1];
+	int			type;
+
+	Assert(expose_recovery || expose_sysid || expose_version);
+
+	do
+	{
+		n = recv(fd, buf, EXPOSE_MAX_QUERY, MSG_PEEK);
+	} while (n < 0 && errno == EINTR);
+
+	/*
+	 * Leave as soon as possible if no chance we are interested. We also
+	 * simply return false for n == -1
+	 */
+	if (n < EXPOSE_MIN_QUERY)
+		return false;
+
+	buf[n] = '\0';
+
+	type = EXPOSE_NOTHING;
+	for (int i = 0; i < lengthof(endpoint_actions); i++)
+	{
+		if (
+			strncmp(buf, endpoint_actions[i].endpoint, strlen(endpoint_actions[i].endpoint)) == 0
+			&&
+			(endpoint_actions[i].require == NULL
+			 ||
+			 *(endpoint_actions[i].require)
+			 ))
+		{
+			type = endpoint_actions[i].type;
+			break;
+		}
+	}
+
+	if (type == EXPOSE_NOTHING)
+		return false;
+
+	{
+		static const char http_version[] = "HTTP/1.1";
+		static const char http_type[] = "Content-Type: text/plain";
+		static const char *http_conn = "Connection: close";
+		static const char http_len[] = "Content-Length";
+
+		StringInfoData msg;
+
+		if (type == EXPOSE_HEAD_REPLICA)
+		{
+			/*
+			 * Caller only cares about the HTTP response code, so no content
+			 * needed
+			 */
+
+			initStringInfoExt(&msg, 64);
+
+			appendStringInfo(&msg,
+							 "%s %s\r\n"
+							 "%s\r\n"
+							 "%s\r\n\r\n",
+							 http_version,
+							 (RecoveryInProgress() ? "200 OK" : "503 Service Unavailable"),
+							 http_type,
+							 http_conn
+				);
+		}
+		else
+		{
+			StringInfoData content;
+
+			initStringInfoExt(&content, 64);
+
+			if (expose_recovery && (type == EXPOSE_GET_ALL || type == EXPOSE_GET_REPLICA))
+				appendStringInfo(&content, "%s%d\r\n",
+								 type == EXPOSE_GET_ALL ? "RECOVERY: " : "",
+								 RecoveryInProgress() ? 1 : 0);
+			if (expose_sysid && (type == EXPOSE_GET_ALL || type == EXPOSE_GET_SYSID))
+				appendStringInfo(&content, "%s%lu\r\n",
+								 type == EXPOSE_GET_ALL ? "SYSID: " : "",
+								 GetSystemIdentifier());
+			if (expose_version && (type == EXPOSE_GET_ALL || type == EXPOSE_GET_VERSION))
+				appendStringInfo(&content, "%s%d\r\n",
+								 type == EXPOSE_GET_ALL ? "VERSION: " : "",
+								 PG_VERSION_NUM);
+
+			initStringInfoExt(&msg, 256);
+
+			appendStringInfo(&msg,
+							 "%s 200 OK\r\n"
+							 "%s\r\n"
+							 "%s: %d\r\n"
+							 "%s\r\n\r\n"
+							 "%s",
+							 http_version,
+							 http_type,
+							 http_len, content.len,
+							 http_conn,
+							 content.data
+				);
+
+			pfree(content.data);
+		}
+
+		do
+		{
+			n = send(fd, msg.data, msg.len, 0);
+		} while (n < 0 && errno == EINTR);
+
+		pfree(msg.data);
+
+		if (n < 0)
+			elog(DEBUG1, "could not send to client: %m");
+
+		return true;
+
+	}
+
+}
diff --git a/src/backend/utils/misc/guc_parameters.dat b/src/backend/utils/misc/guc_parameters.dat
index 6bc6be13d2a..3accf6bec9f 100644
--- a/src/backend/utils/misc/guc_parameters.dat
+++ b/src/backend/utils/misc/guc_parameters.dat
@@ -637,6 +637,24 @@
   boot_val => 'true',
 },
 
+{ name => 'expose_recovery', type => 'bool', context => 'PGC_SIGHUP', group => 'CLIENT_CONN_STATEMENT',
+  short_desc => 'Exposes if the server is in recovery mode without a login.',
+  variable => 'expose_recovery',
+  boot_val => 'false',
+},
+
+{ name => 'expose_sysid', type => 'bool', context => 'PGC_SIGHUP', group => 'CLIENT_CONN_STATEMENT',
+  short_desc => 'Exposes the system identifier without a login.',
+  variable => 'expose_sysid',
+  boot_val => 'false',
+},
+
+{ name => 'expose_version', type => 'bool', context => 'PGC_SIGHUP', group => 'CLIENT_CONN_STATEMENT',
+  short_desc => 'Exposes the server version without a login.',
+  variable => 'expose_version',
+  boot_val => 'false',
+},
+
 { name => 'array_nulls', type => 'bool', context => 'PGC_USERSET', group => 'COMPAT_OPTIONS_PREVIOUS',
   short_desc => 'Enables input of NULL elements in arrays.',
   long_desc => 'When turned on, unquoted NULL in an array input value means a null value; otherwise it is taken literally.',
diff --git a/src/backend/utils/misc/postgresql.conf.sample b/src/backend/utils/misc/postgresql.conf.sample
index c36fcb9ab61..8e8be2cf3af 100644
--- a/src/backend/utils/misc/postgresql.conf.sample
+++ b/src/backend/utils/misc/postgresql.conf.sample
@@ -91,6 +91,14 @@
 					# disconnection while running queries;
 					# 0 for never
 
+
+# - Expose information -
+
+#expose_recovery = off
+#expose_sysid = off
+#expose_version = off
+
+
 # - Authentication -
 
 #authentication_timeout = 1min		# 1s-600s
diff --git a/src/include/postmaster/postmaster.h b/src/include/postmaster/postmaster.h
index 753871071ac..ee1ca2fca36 100644
--- a/src/include/postmaster/postmaster.h
+++ b/src/include/postmaster/postmaster.h
@@ -70,6 +70,10 @@ extern PGDLLIMPORT bool restart_after_crash;
 extern PGDLLIMPORT bool remove_temp_files_after_crash;
 extern PGDLLIMPORT bool send_abort_for_crash;
 extern PGDLLIMPORT bool send_abort_for_kill;
+extern PGDLLIMPORT bool expose_recovery;
+extern PGDLLIMPORT bool expose_sysid;
+extern PGDLLIMPORT bool expose_version;
+
 
 #ifdef WIN32
 extern PGDLLIMPORT HANDLE PostmasterHandle;
diff --git a/src/test/modules/test_misc/meson.build b/src/test/modules/test_misc/meson.build
index 6b1e730bf46..6fe8ac4a9d1 100644
--- a/src/test/modules/test_misc/meson.build
+++ b/src/test/modules/test_misc/meson.build
@@ -17,6 +17,7 @@ tests += {
       't/006_signal_autovacuum.pl',
       't/007_catcache_inval.pl',
       't/008_replslot_single_user.pl',
+      't/009_expose.pl',
     ],
   },
 }
diff --git a/src/test/modules/test_misc/t/009_expose.pl b/src/test/modules/test_misc/t/009_expose.pl
new file mode 100644
index 00000000000..f9363d0bc04
--- /dev/null
+++ b/src/test/modules/test_misc/t/009_expose.pl
@@ -0,0 +1,91 @@
+# Copyright (c) 2025, PostgreSQL Global Development Group
+
+# Test gathering information before authentication via expose_* variables
+
+# Force use of TCP/IP - call before the 'use'
+INIT{ $PostgreSQL::Test::Utils::use_unix_sockets = 0; }
+
+use strict;
+use warnings;
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+use Test::More;
+use LWP::UserAgent;
+
+my $node = PostgreSQL::Test::Cluster->new('node1');
+# Setting as logical here simply to avoid wal_level minimal so we can restart as a replica
+$node->init(allows_streaming => "logical");
+$node->start;
+
+my $server_version = $node->safe_psql('postgres', 'show server_version_num');
+my $bindir = $node->config_data('--bindir');
+my $datadir = $node->data_dir;
+my $cdata = qx{$bindir/pg_controldata -D $datadir 2>&1};
+my ($sysid) = $cdata =~ /Database system identifier:\s+(\d+)/;
+
+my $port = $node->port;
+my $URI="http://localhost:$port";
+my $ua = LWP::UserAgent->new(timeout => 2);
+
+my ($response, $code, $content);
+
+$response = $ua->get("$URI/info");
+$code = $response->code;
+is ($code, 500, "GET /info returns HTTP code 500 when nothing is listening");
+
+$response = $ua->head("$URI/replica");
+$code = $response->code;
+is ($code, 500, "HEAD /replica returns HTTP code 500 when nothing is listening");
+
+$node->append_conf('postgresql.conf', 'expose_recovery=on');
+$node->reload();
+
+$response = $ua->get("$URI/replica");
+$code = $response->code;
+is ($code, 200, "GET /replica returns HTTP code 200 when expose_recovery is on (primary)");
+is ($response->content, "0\r\n", "GET /replica returns '0' when expose_recovery is on (primary)");
+
+$response = $ua->head("$URI/replica");
+$code = $response->code;
+is ($code, 503, "HEAD /info returns HTTP code 503 when expose_recovery is on (primary)");
+
+$response = $ua->get("$URI/info");
+$code = $response->code;
+is ($code, 200, "GET /info returns HTTP code 200 when expose_recovery is on");
+is ($response->content, "RECOVERY: 0\r\n", "GET /info returns 'RECOVERY:0' when expose_recovery is on (primary)");
+
+$node->append_conf('postgresql.conf', 'expose_version=on');
+$node->append_conf('postgresql.conf', 'expose_sysid=on');
+$node->reload();
+
+$response = $ua->get("$URI/info");
+$content = $response->content;
+like ($content, qr/RECOVERY: 0/, "GET /info returns 'RECOVERY: 0' when expose_recovery is on (primary)");
+like ($content, qr/VERSION: $server_version/, "GET /info returns 'VERSION: $server_version' when expose_version is on");
+like ($content, qr/SYSID: $sysid/, "GET /info returns correct SYSID when expose_sysid is on");
+
+$response = $ua->get("$URI/sysid");
+$content = $response->content;
+is ($content, "$sysid\r\n", "GET /sysid returns correct value when expose_sysid is on");
+
+$response = $ua->get("$URI/version");
+$content = $response->content;
+is ($content, "$server_version\r\n", "GET /version returns correct value when expose_version is on");
+
+$node->set_standby_mode();
+$node->restart();
+
+$response = $ua->get("$URI/replica");
+$code = $response->code;
+is ($code, 200, "GET /replica returns HTTP code 200 when expose_recovery is on (replica)");
+is ($response->content, "1\r\n", "GET /replica returns '0' when expose_recovery is on (replica)");
+
+$response = $ua->head("$URI/replica");
+$code = $response->code;
+is ($code, 200, "HEAD /info returns HTTP code 200 when expose_recovery is on (replica)");
+
+$response = $ua->get("$URI/info");
+$content = $response->content;
+like ($content, qr/RECOVERY: 1/, "GET /info returns 'RECOVERY: 1' when expose_recovery is on (replica)");
+
+done_testing();
-- 
2.30.2
